Ask your own question, for FREE!
Mathematics 67 Online
OpenStudy (daniellelovee):

The manager of a small park randomly surveyed residents in the community and asked whether they visited the park in the past year. The table shows the results. What is the 95% confidence interval for the difference between the percent of residents who are younger than 50 and who are 50 and older that visited the park in the past year?

OpenStudy (daniellelovee):

@ganeshie8

OpenStudy (daniellelovee):

is this graph but with different numbers

OpenStudy (daniellelovee):

but idk how to change the number in open study

OpenStudy (daniellelovee):

135..............43 92................48

OpenStudy (daniellelovee):

@zepdrix

OpenStudy (amistre64):

what is 135...43, 92...48?

OpenStudy (daniellelovee):

the numbers that are supposed to be on the chart but idk how to change them

OpenStudy (amistre64):

ok, now how do you usually construct such a confidence interval?

OpenStudy (daniellelovee):

using the difference formula?

OpenStudy (amistre64):

i dont know the name of it, but that does sound like an apt description. something like \[(p_1-p_2)\pm z\left(\sqrt{\frac{p_1q_1}{n_1^3}+\frac{p_2q_2}{n_2^3}}\right)\]

OpenStudy (daniellelovee):

http://www.kean.edu/~fosborne/bstat/06d2pop.html

OpenStudy (amistre64):

yeah, thats the one

OpenStudy (amistre64):

so, where is the difficulty ?

OpenStudy (daniellelovee):

how to find the p^ for the formula

OpenStudy (daniellelovee):

with the numbers from the chart

OpenStudy (amistre64):

number of people in a group, divided by number of old people in all

OpenStudy (amistre64):

p q 135...43 ... (135-43) 92...48 ... ( 92-48) ---- 227 people in total is what im thinking ... but the question seems odd to me. is it asked correctly? no mistakes?

OpenStudy (daniellelovee):

proportions?

OpenStudy (daniellelovee):

and no no mistakes

OpenStudy (amistre64):

135 out of 227 people is the proportion that are younger than 50 92 out of 227 people is the proportion that are 50 and older would you agree?

OpenStudy (daniellelovee):

yes

OpenStudy (daniellelovee):

what next?

OpenStudy (amistre64):

still trying to understand the question .... # surveyed #yes young 135 43 old 92 48 total 227 91 What is the 95% confidence interval for the difference between the percent of residents who are younger than 50 and who are 50 and older that visited the park in the past year? (135-92) +- 1.96 sqrt[ 135*92 /227] seems about the only sensible approach to me, but im unsure about it

OpenStudy (amistre64):

a similar question was asked about a year ago if the time stamp is correct http://openstudy.com/study#/updates/537a329fe4b0e54307961a83

OpenStudy (daniellelovee):

but no one replied

OpenStudy (amistre64):

i see that, but the options presented give us a way to hopefully test the approach. for their work i get 70.96 as an upper limit to the interval, and none of their options has that ....

OpenStudy (daniellelovee):

I have choices too

OpenStudy (daniellelovee):

would it help?

OpenStudy (amistre64):

it would help to see of the method used is at least part of the expected process yes

OpenStudy (daniellelovee):

14.9, 25.7 7.4, 33.2 13.7, 29.6 9.5, 31.1

OpenStudy (amistre64):

there has to be a different approach then, my idea isnt getting us close to an option.

OpenStudy (daniellelovee):

yeah and I've done a lot of stuff to that formula and nothing seems to work :/

OpenStudy (amistre64):

the reason the question seems odd to me is that we are comparing old to young ... but are given how many say yes in each case. and the question is not asking anything about the yes/no portion of it, which would make better sense to me.

OpenStudy (daniellelovee):

EXACTLY thats what im noticing that there is no way to use the survey

OpenStudy (amistre64):

3 options give us an average of 20.3 so the correct answer most likely is: 20.3 +- 1.96(S)

OpenStudy (daniellelovee):

but that is not on the choices i gave you

OpenStudy (amistre64):

it is at best 3 of the choices, whatever the method is spose to be, we most likely need that left part to come out to 20.3 since that is the middle of 3 of the intervals in the options.

OpenStudy (daniellelovee):

I got lost there

OpenStudy (amistre64):

given the 4 options, the average (the middle of the interval) is 20.3 for 3 options. its safe to say that they expect you to get the part of the setup ... and mess up on the 1.96(S) part of it M +- 1.96(S) gives us an interval centered around M in the case of your options, M=20.3 for 3 cases telling us that that portion of the setup is most likely 20.3

OpenStudy (daniellelovee):

im confused im sorry because that makes no sense with the options that I have

OpenStudy (amistre64):

options you gave: 14.9, 25.7 ; midpoint is 20.3 7.4, 33.2 ; midpoint is 20.3 13.7, 29.6 ; midpoint is 21.65 9.5, 31.1 ; midpoint is 20.3

OpenStudy (daniellelovee):

ohhhhhhhhok got it

OpenStudy (amistre64):

a confidence interval will be centered about a point ... M, and then we measure the number of deviations away from M that comprise 95% of the data.

OpenStudy (daniellelovee):

therefore we have to find a formula with the midpoint?

OpenStudy (amistre64):

yes, is there a way to test an option? like do you get a certain number of tries at it?

OpenStudy (daniellelovee):

no just one try :(

OpenStudy (amistre64):

if noone else has a better idea, ill do some reading up on this to see if i can make better sense of it. might take a while tho.

OpenStudy (daniellelovee):

is alright :) would it help if I gave you the link to my book?

OpenStudy (amistre64):

you have a book? lol, yes it would be helpful 20.3 is a percentage ... decimal is .203 43/135 - 48/92 = .203...

OpenStudy (amistre64):

so it appears that the question being asked is about the difference between those who said yes ...

OpenStudy (amistre64):

|43/135 - 48/92| + 1.96sqrt(43(135-43)/135^3+48(92-48)/92^3) gives an upper value of .332, or 33.2

OpenStudy (amistre64):

the way the question is worded is just horrible

OpenStudy (daniellelovee):

so option B then? according to your calculations

OpenStudy (amistre64):

B is the only one that makes any sense to me as far as an answer goes yes

OpenStudy (daniellelovee):

alright thank you very much :) I appreciate your help

OpenStudy (amistre64):

had to work backwards ... the intervals are in percent format, 20.3% is our most likely midpoint from the options .203 is the decimal format ... we can obtain that from subtracting the yes/age proportions, which means that our standard error calculations for the sqrt part uses the complements of those proportions. as such, .332 or 33.2% is the high end of our 95% interval (z=1.96)

OpenStudy (amistre64):

is it correct? i have no idea lol .... but it is the only one that makes sense to me as far as the options go. the wording of the question itself remains a mystery to me.

OpenStudy (daniellelovee):

it was correct thank you for your help :)

OpenStudy (amistre64):

yay!! :) good luck

Can't find your answer? Make a FREE account and ask your own questions, OR help others and earn volunteer hours!

Join our real-time social learning platform and learn together with your friends!
Can't find your answer? Make a FREE account and ask your own questions, OR help others and earn volunteer hours!

Join our real-time social learning platform and learn together with your friends!